A handcrafted, untrained linear neural network encodes several integers into one floating-point number by digit concatenation and decodes them using hardware-level bit truncation.
Can neural networks do arithmetic? a survey on the elementary nu- merical skills of state-of-the-art deep learning models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.NE 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
A Synthetic Pseudo-Autoencoder Invites Examination of Tacit Assumptions in Neural Network Design
A handcrafted, untrained linear neural network encodes several integers into one floating-point number by digit concatenation and decodes them using hardware-level bit truncation.